Horizontal Collaboration for Sustainable Transport and Logistics

HAL Id: tel-01665920

https://hal-mines-paristech.archives-ouvertes.fr/tel-01665920

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Logistics

Shenle Pan

To cite this version:

Shenle Pan. Horizontal Collaboration for Sustainable Transport and Logistics. Business administra-tion. Université de Valenciennes et du Hainaut-Cambrésis, 2017. �tel-01665920�

Habilitation à diriger des recherches

UNIVERSITE DE VALENCIENNES ET DU HAINAUT-CAMBRESIS

Sciences Pour l’Ingénieur (SPI) Equipe de recherche, Laboratoire : LAMIH UMR CNRS 8201

Horizontal Collaboration for Sustainable Transport and

Logistics

JURY

- Mme Macharis, Cathy. Professeur. Vrije Universiteit Brussel. Rapporteurs

- Mme Thierry, Caroline. Professeur. Université Toulouse Jean Jaurès. - M. Grabot, Bernard. Professeur. Ecole Nationale d'Ingénieur de Tarbes. - M. Frein, Yannick. Professeur. Grenoble INP.

Examinateurs

- M. Ballot, Eric. Professeur. MINES ParisTech, PSL-Research University. Directeur de Recherche

Acknowledgements

This Habilitation thesis could be considered as the second breakthrough during my research career, the first one would be my doctoral thesis during 2007-2010. In the last ten years, time has passed so fast to me, because I have done my research which is ‘a little bit’ more complicated than I imagined. But I feel very lucky that I have been working with the people who I admire and appreciate. I am very grateful for their precious help.

I would like to express my deepest gratitude to Dr. Eric Ballot, Professor at MINES ParisTech. I have been working with Eric since 2007 when he was the principal supervisor for my doctorate. And he has always been the most important, reliable, trustworthy mentor for my academic career. I appreciate a lot his continued support to young researchers, his great personality, and his rigorous way of working. I have learnt a lot from him and I am very thankful to him.

I would also like to express my sincere gratitude to Dr. Damien Trentesaux, Professor at Université de Valenciennes et du Hainaut-Cambrésis. This Habilitation thesis would not be possible without his support and help. His suggestions and guidance were preciously valuable for this thesis.

I would also like to sincerely thank the members of examining committee, Dr. Caroline Thierry (Professor at Université Toulouse Jean Jaurès), Dr. Bernard Grabot (Professor at Ecole Nationale d'Ingénieur de Tarbes), Dr. Yannick Frein (Professor at Grenoble INP), and Dr. Cathy Macharis (Professor at Vrije Universiteit Brussel). I am very grateful for the precious, constructive guidance and suggestions that they provided to my research works. These comments are extremely helpful to my future research.

I would also like to thank my colleagues at MINES ParisTech. I am very grateful to work in such a great university and research centre, and work with these amazing people. I would like to address my acknowledgement to all the colleagues at the Centre de Gestion Scientifique for their valuable supports, Pascal, Benoit, Blanche, Franck, Sophie, Cédric, Kevin, Philippe and all others, and specially to our lovely secretaries Céline, Stéphanie, Marie-Michelle, and my teammates in our research group of Chair Physical Internet.

I would like to express my special thanks to everyone who has helped me with my research, for their valuable suggestions and thoughtful help during my career, particularly, Frédéric Fontane (MINES ParisTech) who was co-supervisor for my doctorate, Benoit Montreuil (Georgia Tech), George Huang (University of Hongkong), Vaggelis Giannikas (University of Bath, UK), Duncan McFarlane (University of Cambridge, UK), Yves Sallez (Université de Valenciennes, France), Patrick Charpentier and Nicolas Krommenacker (ENSTIB, France), Ray Zhong (University of Auckland, New New Zealand), Alexandra Dolgui (MINES Nantes, France), Chao Chen (University of Chongqing, China), Yufei Han (Symantec, France), Remy Glardon (who is enjoying his new life).

Last but not least, I would like to express my utmost gratitude to my family, my wife Haley, my parents, my elder brother and his family. Their constant support of my work and endless love make my life easier and happier.

Foreword

This document presents only the scientific part of the Habilitation thesis of Dr. Shenle Pan, that is defended on 1st December 2017, at Université de Valenciennes et du Hainaut-Cambrésis.

The Habilitation thesis (also called post-doc thesis) has two purposes. The first is to obtain the degree of Habilitation à Diriger des Recherches (HDR) in France, which is a degree of accreditation to supervise research. Knowing that it is required for supervising doctoral research (PhD students) and applying for professorship in France. Second, the thesis tends to investigate the research area to which I have been devoted for ten years, i.e. horizontal collaboration in transport and logistics. For that purpose, the thesis will review the recent research works and the state of the art of the research area. Then, my scientific contributions as well as my research program for the next research lines will be discussed under a such context, in order to demonstrate my capacity of supervising and leading doctoral research. The Habilitation thesis comprises two parts, the first part concerned with my Curriculum Vitae, and the second part devoted to the scientific content of the thesis. With regard to the second part, which is presented in this document, it focuses on recent advances of research on the horizontal collaboration in logistics for sustainable transport and logistics operations. It investigates also scenarios of the future logistics system.

To my family

Table of Contents

GENERAL INTRODUCTION ... 11

LIST OF ABBREVIATIONS ... 13

CHAPTER 1 LOGISTICS: SUSTAINABILITY AND EVOLUTION ... 15

1.1 Chapter Introduction ... 15 1.2 Fundamentals of Logistics ... 15 1.3 Sustainability in Logistics ... 18 1.4 The Evolution of Logistics Organisations ... 22 1.5 Limitations and Scope of This Habilitation Thesis ... 29 CHAPTER 2 HORIZONTAL COLLABORATIVE AND INTERCONNECTED LOGISTICS: STATE OF THE ART ... 31

2.1 Chapter Introduction ... 31 2.2 Horizontal Collaborative Transport and Sustainability ... 31 2.3 Analysis of Research Contributions ... 32 2.4 Research Trends and Gaps ... 43 2.5 Research Prospects ... 44 2.6 Conclusion of the Chapter ... 46 CHAPTER 3 MY SCIENTIFIC CONTRIBUTIONS TO HORIZONTAL COLLABORATIVE AND INTERCONNECTED LOGISTICS IN FMCG SECTOR ... 47 3.1 Chapter Introduction ... 47 3.2 Contributions to Logistics Pooling ... 48 3.3 Contributions to Physical Internet ... 58 3.4 Conclusion of the Chapter ... 77 CHAPTER 4 TOWARDS A NEW ERA OF OPEN, INTELLIGENT, DECENTRALISED LOGISTICS SYSTEMS ... 79

4.1 Chapter Introduction ... 79 4.2 Context and Motivation ... 79 4.3 A Brief Review of the SoLS-related Literature ... 80 4.4 Functionalities of SoLS ... 83 4.5 Research Prospects ... 87 4.6 Conclusion of the Chapter ... 93 CHAPTER 5 ... 95 GENERAL CONCLUSION ... 95

APPENDIX A: SUMMARY AND POSITIONING OF MY RESEARCH WORKS UNDER THE FRAMEWORK OF HCT SOLUTIONS AND IMPLEMENTATION ISSUES ... ERROR! BOOKMARK NOT DEFINED. REFERENCE LIST ... 97

G

ENERAL

I

NTRODUCTION

This Habilitation thesis aims to investigate the research area of horizontal collaboration in transport and logistics. It will focus on the state of the art, and the trends and future research lines of the research area. Moreover, we will also discuss my research contributions to the area since 2007, and my short-term and long-term research program related to the area.

This Habilitation thesis consisting of 5 chapters is organised as follows:

§ Chapter 1 will introduce fundamentals, objectives and challenges, and organisations of logistics. More precisely, the objective of this chapter is threefold: first, to give a consensus definition of logistics and demonstrate its importance, especially with regard to sustainability; second, to summarise and analyse the past, undergoing and emerging evolutions in logistics organisations, then to specify the scope of analysis in this thesis; third, to briefly introduce the scientific issues under the scope that will be discussed in the rest of the part.

§ Chapter 2 focuses on the organisation of horizontal collaborative and interconnected logistics (see Type 3 in Figure 6, Chapter 1). An exhaustive survey of all horizontal collaborative transport solutions in the literature, as well as the relevant scientific issues, will be presented. The objective of this chapter is to identify, analyse and position the major international and national contributors to this topic.

§ Chapter 3 introduces the research contribution of my own research works since 2007, and the doctoral theses that I have been co-supervising since 2010. The contribution will be discussed solution by solution and issue by issue under the research framework proposed in Chapter 2, in order to position my contributions onto the international and national research community.

§ Chapter 4 focusing on the organisation of open, intelligent, decentralised logistics (see Type 4 in Figure 6, Chapter 1). It deals with research prospects for short terms and for long term. First, we will give a theoretical framework of such logistics system, in order to outline its profile. The essential functionalities and requirements, as well as relevant scientific issues will also be discussed. Then, we will give some research lines toward such system, coupling with some research topics (ex supervising or developing Ph.D. theses). The objective of this chapter is to draw a framework for my next research topics.

L

IST OF ABBREVIATIONS

1PL – First-Party Logistics, see also 2PL, 3PL, etc. AR/VR – Augmented Reality/Virtual Reality

B2C – Business-to-Customer, see also B2B, B2B2C CLCP – Collaborative Lane Covering Problem

CPFR – Collaborative Planning Forecasting and Replenishment CRC – Collaborative Routing Centre

CVRP – Collaborative Vehicle Routing Problem, see also VRP DC – Distribution Centre

ECR – Efficient Consumer Response EDI – Electronic Data Interchange FCE – Flow-Controlling Entity FMCG – Fast Moving Consumer Goods

FTL/LTL – Full-Truckload/Less-than-Truckload, and TL equal to FTL HC/VC – Horizontal Collaboration/Vertical Collaboration

HCT – Horizontal Collaborative Transport HDV – Heavy Duty Vehicle

HMAS – Holonic Multi-agent System, see also MAS ICT – Information and Communications Technology IoT – Internet of Things

IT – Information Technology JRP – Joint Route Planning KPI – Key Performance Indicator LLC – Limited Liability Company LSP – Logistics Service Providers

MILP – Mixed Integer Linear Programming OTC – Open Tracing Container

PI – Physical Internet PoS – Point of Sale

RFID – Radio-Frequency Identification RTI – Returnable Transport Item SC – Supply Chain

SCM – Supply Chain Management SME – Small-Medium Enterprise

SoLS – Self-organising Logistics System SV – Shapley Value

TSP – Transport Service Procurement UCC – Urban Consolidation Centre VMI – Vendor Managed Inventory WDP – Winner Determination Problem WH – Warehouse

C

HAPTER

1

Logistics: Sustainability and Evolution

1.1 Chapter Introduction

This introductory chapter comprises 5 sections. Section 2 presents some fundamentals of logistics and establishes some consensus definitions which will be used throughout the thesis. Section 3 discusses the importance of logistics, especially with regard to sustainability. Section 4 broadly introduces the main challenges and objectives in logistics today, and consequently past, current, and emerging logistics organisations. Finally, Section 5 indicates the scope and limitation of this thesis.

1.2 Fundamentals of Logistics

A variety of definitions exist so it is necessary to provide a widely recognised concept of logistics. Some basic aspects such as definition, activities, and scope are discussed and these concepts will be adopted throughout the rest of this thesis.

1.2.1 General definition and activities

Logistics is a broad discipline in Management Science that consists of many activities. In order to clearly define the scope of this document, it is necessary to define logistics and the activities involved.

A number of definitions can be found in the literature, see Lambert and Cooper (2000), Chopra and Meindl (2004), Christopher (2005), Harrison and van Hoek (2005) and Ballou (2007) for example. In this thesis, we have selected and use the most up-to-date, common definition. Generally speaking, logistics has been considered as a part of Supply Chain Management (SCM) and can be defined as the management of goods including planning, implementing, and controlling the forward and reverse flows from the production point to the final consumption points to satisfy end-consumer demand. This definition is adapted from that given by the Council of Supply Chain Management Professionals (CSCMP, 2017).

Figure 1. Activities associated with logistics and supply chain management (Ballou, 2007) More specifically, logistics involves a number of activities from the manufacture (at plant level) to the distribution of goods (at chain or network level). The example presented in Figure 1 adapted from Ballou (2007) defines the scope of logistics and its activities. Generally speaking, logistics focuses on the management of physical and information flows, without compounding IT services, marketing or finance which are usually considered at supply chain management level. Hereinafter, we will use this definition of logistics.

1.2.2 Scope: from individual activity to networks

The meaning and scope of “logistics” has evolved over the past few decades (Jahre and Fabbe-Costes, 2005). Initially, logistics represented the management of separate activities such as storage and transport to move physical goods. Then, some began to reorganise and integrate the separate activities to achieve better, more global optimisation. Building on these ideas, companies then introduced flow into logistics and organised the related activities as a whole. On a production site, for example, inbound flows involved moving materials from the supplier’s site to the production site, internal flows concerned on-site production logistics, and outbound flows concerned distribution from the site to customers. Subsequently, companies working together extended logistics to the supply chain with the aim of strengthening coordination and inter-management between companies to ensure global, mutual optimisation. The vision of logistics was thus extended from a single company to a chain of companies. Information flow management plays a vital role in coordinating and managing physical flows in such multi-company organisations, which is why information and physical flows are usually considered together in logistics. Accordingly, the power of “logistics” has been recognised step-by-step, from operational to strategic level.

In this thesis, we consider the most recent vision taking into account multiple chains, namely a logistics network. Since 2000, the term supply network has been used to define supply chain management in a multiple supply chain and multi-party network (Choi et al., 2001; Harland,

1996; Harland et al., 2001; Jahre and Fabbe-Costes, 2005). The objective is to organise the activities within the network (logistics and supply chain management, as shown in Figure 1) and optimise them as a whole by crossing supply chain borders. The notion of supply network has recently been extended from a set of supply chains to a set of networks, i.e. a network of networks. For example, the concept of logistics web (or supply web) was introduced by Hakimi (2014) and is defined as “a network of interrelated supply networks, each embedding interlaced supply chains involving multiple organisations with collaborative or competitive supply relationships”. Montreuil (2011) and Ballot et al. (2014) propose a similar idea with the concept of the Physical Internet (PI), which is defined as a network of independent networks. The idea is, as a metaphor of the digital internet, to create an open, shared network to interconnect independent logistics networks and services. All resources within the network (distribution centres, assets, information systems, etc.) are standardised and interconnected to provide high interoperability between companies for the optimal use and sharing of a common network – the PI.

In the literature, the difference between supply network and logistics network is similar to that between supply chain management and logistics, as illustrated in Figure 1. However, the difference is inadequately discussed in the literature, especially due to the lack of definition for logistics network. Knowing that in this thesis we will focus on logistics and then logistics networks, a broad, consensus definition is necessary. We use the term “logistics network” to broadly cover all network-based organisations of logistics activities. Based on the studies discussed above, the following definition is proposed and used in this thesis: a logistics network consists of a number of companies (manufacturers, forwarders, distributors, service providers, etc.), their physical sites (warehouses, distribution centres, hubs, etc.), physical assets (goods, transport means, pallets, etc.), and working agreements (contracts, charters, rules, etc.), of which the aim is to optimise and mutualise the organisation and management of logistics activities involving physical and information flows within the network as a whole to achieve synergy. On the basis of this definition, we focus on the management of physical and information flows within a network, without considering other aspects such as finance or marketing.

1.2.3 The objectives of logistics

Logistics should be able to deal with increasing demands for goods while satisfying challenging logistical constraints such as demand uncertainty, lead time, and resource availability. Consequently, logistics organisations are nowadays expected to be efficient, effective, and responsive while respecting other objectives such as sustainability and resilience. Table 1 summarizes the main objectives observed in the literature.

Objectives Short Description

Effectiveness How well a goal is adequately met (Mentzer and Konrad, 1991)

Efficiency How well the resources expended are utilised to meet a goal (Mentzer and Konrad, 1991)

Agility Ability of a system to rapidly reconfigure (Bernardes and Hanna, 2009)

Flexibility Ability of a system to change status within an existing configuration (Bernardes and Hanna, 2009)

Resilience Ability of a system to return to its original state or move to a new, more desirable state after being disturbed (Bhamra et al., 2011)

Sustainability Ability of a system to protect, sustain and enhance the human and natural resources that will be needed in the future while meeting a desired goal (Labuschagne et al., 2005)

The five objectives shown in Table 1 can also be considered as criteria to assess the performance of a logistics system. For example, a logistics system may be effective (e.g., all customers are very satisfied) but inefficient (e.g., cost to satisfy customers is too high). Simultaneously meeting all objectives, i.e. good performance on all criteria, is an obvious challenge in logistics today.

Both my research and this thesis focus on sustainability. However, it is worth recalling the objectives in order to further expose the fact that to simultaneously respect sustainability and other criteria is a fundamental challenge for today’s logistics systems. We are, therefore, looking for new, innovative organisations to provide solutions. The next part will further explain why it is worth studying sustainability in logistics.

1.3 Sustainability in Logistics

It has been recognised for many years that logistics plays a significant role in economic growth and social development. After 2000 in particular, the environmental impact due to logistics activities such as CO2 emissions, land use, truck noise, and traffic congestion, also

became a major concern. Accordingly, sustainable development has gained much attention from both researchers and practitioners in logistics. In this part, we discuss the significance of logistics according to three aspects: economic impact, environmental impact, and social impact - the three pillars of sustainable development (European Commission, 2001) - in order to illustrate the motivations behind and the importance of the research.

1.3.1 Logistics and economy

The economic importance of logistics can be justified by its cost as a percentage of GDP (Gross domestic product). According to Figure 2, logistics costs in 2013 (mainly transport and inventory) represented approximately 13% of the GDP in Europe and 17% in Asia. Moreover, growth in GPD is correlated with growth in logistics activities, especially freight transport. For several decades, freight transport has seen rapid, sustained growth generally coupled with growth in GDP, as shown in Figure 3. Despite having observed the decoupling of road freight transport and economic growth in some countries, as in the UK over the past decade (McKinnon, 2007), we can assume that freight transport and logistics will remain an important part of the European economy.

As stated in a report issued by the French government (Savy, 2015), logistics costs contributed an estimated 10% to the national GDP, which represents over €200 billion. In terms of volume, inland freight transport in France (excluding pipeline) was an estimated 328 Million tkm (tonne-kilometres) in 2013, of which road transport accounted for almost 88%.

Figure 2. Logistics costs as a percentage of GPD (source: http://www.scdigest.com1_{, figure adapted}

from 2013 CSCMP ‘State of Logistics’ Report)

Figure 3. Growth of GDP and freight transport in billion tkm (from European Environment Agency: positive values in the columns indicate higher growth in GDP, while negative values indicate higher

growth in freight transport; the data refer to road, rail, and inland waterway transport).

These data show that logistics and freight transport are an important sector for the economy and its growth. ALICE (ALICE, 2016), the European Technology Platform on Logistics, estimated that “a 10% to 30% improvement in efficiency in the EU logistics sector would potentially equal a €100-300 billion cost relief for the European industry”. In other words, EU logistics is a huge market with significant potential savings.

Logistics also plays an essential role in the economic performance of businesses. Take the e-commerce giant Amazon, for example. Over the past few years, the company has been increasingly recognized as a leading innovator in logistics worldwide. However, one of the pressures came from financial losses on outbound shipping logistics, see Figure 4. From this figure, we can clearly see that Amazon was losing more and more money on outbound logistics, losing $5 billion on shipping costs in 2015. However, thanks to the efforts on innovation, the increase in losses was reduced. The same results can also be observed in other

EEA-33 excluding Croatia and Liechtenstein — Freight transport volumes and GDP

tkm GDP Decoupling 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 _{2011 2012*} -15 -10 -5 0 5 10 15 70 80 90 100 110 120 130 D eco u pl in g ( % ) T ra n spo rt vo lu m e an d G D P in de x (2 0 0 5 = 1 0 0 )

companies and sectors such as manufacturing, traditional retailing, and logistics service providers.

Figure 4. Amazon’s shipping revenue and costs 2006-2015 (source Statista.com) Hence, logistics is of interest to both society and companies in terms of the economy.

1.3.2 Logistics and the environment

One of the major environmental issues of freight transport is greenhouse gas emissions, especially CO2 (Nouira et al., 2016). In many countries, the freight transport sector is an

important contributor to CO2 emissions. For example, road freight traffic accounts for

approximately 6% of total domestic CO2 emissions in the UK (Piecyk and McKinnon, 2010)

and approximately 9% in France (ADEME, 2007). According to Savy (2015), in France in 2012, road freight transport alone produced approximately 33 Mt CO2. More globally, the

World Economic Forum and Accenture reported in 2009 that logistics activities, including freight transport and logistics buildings, accounted for approximately 5.5% of global greenhouse gas emissions (McKinnon et al., 2012). Despite the different time frames and scope, these figures show that the freight transport sector contributes significantly to CO2

emissions and should, therefore, be aware of the need to reduce emissions.

Some well-known objectives have been established for reducing CO2 emissions. For example,

according to the low-carbon economy roadmap established by the European Commission2, the EU must cut CO2 emissions to 20% below 1990 levels by 2020, 40% by 2030, and 60% by

2040, to reach 80% by 2050. All sectors must contribute to these objectives, especially the transport sector, including freight transport. From Figure 5 we can see that since 1990, total emissions have only been increasing in the transport sector. The EU 2050 target in this sector is obviously more ambitious compared to the other sectors. For example, the EU roadmap (Figure 5) envisages reducing emissions from transport to more than 60% below 1990 levels by 2050. Therefore, radical improvements in efficiency in this sector, especially in freight transport, will be required to meet this target.

-6 000 -4 000 -2 000 0 2 000 4 000 6 000 8 000 10 000 12 000 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Amazon's outbound shipping from 2006 to 2015 (in million U.S. dollars)

Figure 5. EU GHG emissions towards an 80% domestic reduction by 2050 (100% =1990) (source: http://ec.europa.eu/clima/policies/strategies/2050/index_en.htm)

In addition to emissions, other environmental issues such as energy consumption, noise, land use, and waste are also associated with freight transport and logistics (Goldsby et al., 2014; Korzhenevych et al., 2014; Taniguchi and Thompson, 2014). All these problems can generally be seen as the negative externalities of logistics. These externalities can be described as external costs that are the costs to society but are not taken into account by transport users unless there are policies dictating otherwise (Maibach et al., 2008). In addition, these external costs are often neglected as they are difficult to justify quantitatively. However, thanks to some econometric approaches, see the handbooks (Korzhenevych et al., 2014; Maibach et al., 2008) for example, the external costs of freight transport can be monetarily estimated, especially external environmental costs. The environmental impact of logistics on our daily lives is thus more concrete and intuitive. Hence, means of reducing all these negative externalities is becoming a major concern in modern logistics for both service providers and users.

1.3.3 Logistics and society

Logistics has both positive and negative impacts on society. Logistics has obviously improved our daily life. For example, effective and efficient logistics can help traditional retailers to reduce logistics costs so as to provide a wide diversity of goods at a low price for end consumers. Shopping is also becoming much easier thanks to recent multi-retailing channels, e.g., online stores, mobile stores, which also rely on strong logistics support. As a result, in China, for instance, the number of packages delivered across the country in 2016 was over 30 billion, which is 53% more than 2015 (National Bureau of Statistics of China). Logistics also contributes to the job market. According to the French government report by Savy (2015) and some statistic reports by INSEE3, logistics, including transport and warehousing, created 1.3 million jobs in France in 2012, which is approximately 5.7% of all employees within the French economy (all jobs included). These examples show that we all have more or less benefited from the development of logistics.

However, logistics also has a negative impact on society, especially with regards to safety, working conditions, and traffic congestion. According to a report issued by the INRS in France4, employees in the logistics sector are injured 2 to 3 times more often than employees in other sectors. According to a French government report issued by the SOeS (2013), 2 946

heavy-duty truck accidents occurred in France in 2012 and 486 lives were lost. The same report also indicates that heavy-duty truck drivers in France work more than 45 hours per week on average and spend 65% to 75% of their time driving. Such indecent working conditions may eventually lead to a higher risk of traffic accidents and this must be improved. The contribution of freight transport to traffic congestion is another important concern for our daily lives, either for inter-city or intra-city traffic, as it may increase our waiting time, driving time, and operating costs. In terms of the economy, the handbook in Maibach et al. (2008) (updated by Korzhenevych et al. (2014)) proposes econometric approaches to estimate the external societal costs of traffic accidents and congestion, which may help us to understand the significance of freight transport and logistics from a social perspective.

Supported by the aforementioned figures, it is clear that logistics is of critical significance to society. Improving the efficiency of logistics within the context of sustainable development is a common goal for industrialists, consumers, and policy makers. The main potential for improving logistics efficiency can be found at organisational level rather than operational level. It is in this context that both researchers and practitioners have been looking for more efficient and effective logistics organisations and solutions for many years. The next section presents how logistics organisations have been becoming more sustainable.

1.4 The Evolution of Logistics Organisations

To understand better how logistics organisations have evolved over the past few decades, this section reviews the current logistics organisations. Four organisations are discussed in chronological order. For each type, we will discuss motivations, enablers, main stakeholders, and key features. Figure 6 illustrates the different organisations.

Figure 6. Evolutions of logistics organisations

Logistics Organisation 2PL/3PL 4PL Open & intelligent network

From the beginning

Type 1 In-house logistics After 70’s Type 2 Outsourcing After 2005 Type 3 Horizontal Collaboration & interconnection After 2010 Type 4 Open, intelligent and decentralised Type 1PL 5PL Short Description

Logistics is organised and fulfilled by shipper – manufacturer (or receiver – client).

Logistics is outsourced to and fulfilled by professional third party service providers. The logistics performance relies on vertical collaboration between shipper – 3PL – receiver.

Horizontally cross-chain collaboration that is to share resources, co-define logistics plans, or co-design networks, in order to mutually improve logistics performance for multi-players and chains.

An open logistics system that is smart, agile, and able to collect and analyse real-time information to make real-time decisions (resources planning, product recall, etc.).

Table 2 specifies the LSP terms (logistics service providers) used in Figure 6. Despite there being a variety of definitions for 1PL to 4PL in the literature, we have adopted the most widely accepted ones given in the thesis by Cruijssen (2006). However, there is no consensus definition for 5PL. The definition given in Hosie et al. (2012) is cited as it is similar to the idea in this thesis, but in this thesis we do not aim to discuss the concepts and definitions of LSPs in detail. Here, we only cite the definitions from the literature that will help us to clarify the ideas that will be presented in the next section.

LSP Description

1PL In a 1PL concept, logistics activities are not outsourced, but performed in-house by the shipper. 2PL In the 2PL concept, a shipper outsources transport to a carrier company that is expected to perform a number of clear-cut tasks. The planning and organisation remain in the hands of the shipper.

3PL A 3PL (third party logistics service provider) allows shippers to outsource a whole package of logistics services. This LSP takes the responsibility for the planning and organization and in that role communicates with both the shipper and the receiver(s) of the goods.

4PL A 4PL concept represents a situation where even the management of logistics activities is outsourced. The 4PL focuses entirely on this management task and therefore generally does not own logistics assets. This concept becomes beneficial if the 4PL manages multiple supply chains amongst which synergies can be exploited.

5PL A 5PL provider manages at the strategic level by a focus on providing innovative logistics solutions throughout the entire supply chain (or supply network). The major tasks of 5PL companies include mapping and reengineering the supply chain, the 4PL functions (integration and control of transport, handling, warehousing, etc.,) and providing integrated information systems to ensure real-time visibility and control of the entire supply chain. 5PL providers are almost fully virtual.

Table 2. Concept and categories of LSP (1PL to 4PL cited from Cruijssen (2006), and 5PL cited from Hosie et al. (2012))

Below, each type is discussed in succession. Figure 6, supported by Table 2, can be seen as a guiding tool for the discussion.

1.4.1 Type 1 – In-house logistics

Initially, logistics was planned, executed, and controlled by shippers using privately owned resources (including creating own logistics company). This so-called in-house logistics is also called 1PL (first-party logistics) since only one party - shipper is involved in this organisation. As the primary objective of logistics is customer satisfaction through reliable services, i.e. quality, dependable services, early in-house logistics relied on direct shipment to customers, for example from the shipper’s factory (or warehouse) to the customer’s address. The main issue with this organisation is the significant increase in logistics costs when customer demands and numbers increase. Shippers are therefore incited to procure logistics services from external providers if they can provide good service at a low price in order to decrease logistics costs.

Nowadays, many companies have maintained their logistics in-house. Some well-known examples can be found in the automobile sector (Manners-Bell, 2014). One of the main reasons is to ensure the reliability of deliveries to customers. For some shippers, keeping logistics in-house may result in better customer service and lower costs than outsourcing5,6. In

reality, outsourcing logistics or not, i.e. Make or Buy strategy, is not always obvious, see Abdur Razzaque and Chang (1998), Cánez et al. (2000), Wilding and Juriado (2004) and Kremic et al. (2006). But this question is not within the scope of this thesis.

1.4.2 Type 2 – Outsourced logistics

Once companies (e.g., manufacturers) decide to procure logistics services from external companies who are usually professional logistics service providers, the logistics is outsourced to 3PLs (Third-party logistics providers) (Selviaridis and Spring, 2007). Sometimes the term 2PL is also used to differentiate the degree of outsourcing. Generally speaking, 2PLs can refer to carriers who execute the transport activity without planning or controlling the distribution (inventory control in particular), while 3PLs will ensure all the logistics activities including transport, warehousing, inventory control, distribution planning, etc. Both of them are asset-based companies so as to be able to execute the outsourced logistics activities. Hereinafter they are called 3PL.

At this stage, 3PLs struggled to reduce logistics costs to propose cost-efficient solutions to their clients (shippers) while maintaining effective logistics services. The pressure on logistics costs was thus partially shifted from the shipper (e.g. manufacturing industry) to the 3PL, especially for international 3PLs as a result of globalisation. To this end, 3PLs soon recognised that it was important to collaborate with shippers and receivers. However, shippers and receivers were also incited to collaborate with 3PLs as they would also benefit from lower prices or better services, for example. This is called vertical collaboration in a supply chain (Mason et al., 2007). ECR (Efficient Consumer Response) is a measure based on this initiative to optimise the supply chain: “ECR seeks to optimize the (grocery) supply chain, minimizing inventory levels and optimizing product availability. By taking a holistic approach, time and cost can be stripped from the supply chain. It also addresses product quality”, defined in Wood (1993). ECR then became a joint trade and industry body in many countries and is particularly active in the FMCG (fast moving consumer goods) sector, see ECR France7, for example. More specifically, companies in a vertical relationship (e.g., suppliers, manufacturers, LSP, retailers) aim to collaborate to build an integrated supply chain (Harrison and van Hoek, 2005; Jayaram and Tan, 2010; Stank and Goldsby, 2000). Well-known solutions developed for this purpose include VMI (vendor managed inventory) models (Marquès et al., 2010; Yao et al., 2007), collaborative transport planning (Stadtler, 2009), and CPFR (Collaborative Planning, Forecasting and Replenishment) (Holmström et al., 2002; Lehoux et al., 2011).

1.4.3 Type 3 – Horizontal collaborative and interconnected logistics

Since the 90’s, radical changes have been made to traditional supply chains as a result of new manufacturing and retailing practices aimed at reducing inventory and improving efficiency throughout the entire chain (Dornier, 1997; Pan, 2010). Upstream, LEAN manufacturing practices such as Just-in-time and delayed differentiation were popularized in the manufacturing industry. While considerably reducing stocks in the chain (Tavasszy et al., 2012), these practices upstream (e.g., automotive suppliers) resulted in increasingly demanding logistical constraints – small and accurate shipment size with low delay tolerance. Downstream, retailers strived to reduce inventory levels (both at points of sale and at distribution centres (DC)) by accelerating the inventory turnover rate (Chopra and Meindl, 2004). In the meantime, points of sale (PoS) in cities had also been downsized in order to get closer to end consumers. As a result, shipments to DC and PoS were considerably downsized

but with shorter delivery times and higher frequencies, see the reports from France (IFSTTAR, 2013) and the U.S. (Bureau of Transportation Statistics, 2006). All these changes resulted in fragmented and accelerated freight flows (Ballot and Fontane, 2010).

Faced with the fragmentation and acceleration of freight flows, better responsiveness and flexibility were necessary in logistics with regard to individual customers. A responsive, flexible logistics system should be able to quickly respond to consumer needs and quickly adjust its status in the current organisation to overcome a diversity of logistical constraints and difficulties, see Bernardes and Hanna (2009), Wilding et al. (2012) and Jafari (2015). In this context, traditional 3PL-based outsourcing was challenged by the fact that it was increasingly difficult to optimise transport for fragmented and accelerated freight flows, and thus inadequate performance could be observed. Considering that transport-related activities can represent more than 60% of total logistics costs (Goldsby et al., 2014), the inefficiencies observed are of critical significance: average truck fill-rates by weight were 60% to 70% and approximately 20% to 25% of trucks were running empty (Department For Transport, 2008; Léonardi and Baumgartner, 2004; McKinnon et al., 2003; PIPAME and CNAM, 2009). As a result, some manufacturing companies quit outsourcing and returned to in-house logistics8. Furthermore, logistics-related inefficiencies are also an issue with regard to sustainable development bearing in mind the sustainability challenges discussed in the previous section and the inefficiencies of logistics. Considering all this, we naturally ask the question whether traditional 3PL-based outsourcing is sustainable in the face of these new challenges.

It is in this context that research on logistics horizontal collaboration and interconnection of logistics networks began in 2000. Logistics horizontal collaboration describes cross-chain collaboration between two or more firms that operate at the same level of the supply chain to create logistics synergies (Cruijssen, 2006). Examples include lane or request exchange between carriers (Özener et al., 2011), joint route planning between shippers (Cruijssen et al., 2007a), and logistics pooling between SC (Pan et al., 2013). The Physical Internet, another concept of horizontal collaboration, has more recently been introduced and aims to interconnect and integrate logistics networks that are currently independently managed and controlled by different entities based on the modularisation and standardisation of processes and materials within a common network (Ballot et al., 2014). All these concepts and solutions will be discussed exhaustively in the next chapter.

This type of organisation exploits cross-chain or cross-network logistics synergies by sharing logistics resources (means of transport, warehouses, distribution channels, etc.). To this end, a new organisation is necessary, namely 4PL. A 4PL can be described as a non-asset-based company whose job is to manage logistics activities (and assets) for one or several companies, for example, manufacturers or retailers (Cruijssen, 2006). A 4PL or a group of 4PL can coordinate flows of different chains or networks so as to exploit logistics synergies. It can also be called a “trustee”, with the additional duty of allocating gain to partners (see Vanovermeire and Sörensen (2014a), and the Collaboration Concepts for CO-modality (CO3) project discussed in Rossi (2012)). A practical example includes the concept of cross-chain collaboration centre (4C). In reality, this is a control tower for managing, executing, and controlling cross-chain logistics activities proposed in Kok et al. (2015). Examples of businesses include TRI-VIZOR located in Belgium, claimed as the world’s first cross-supply chain orchestrator (TRIVIZOR, 2016); or CRC® Services located in France, which is a pooled

cross-docking platform managed by an independent 4PL for manufacturers and multi-distributors in the FMCG sector (CRC, 2016).

1.4.4 Type 4 – Open, intelligent, decentralised logistics

After 2010, we can observe that customisation and personalisation has become the trend in production and in logistics. It leads to nowadays logistics that is characterised with high deliver frequency, short lead-time, low volume, and multi-distribution channel. Accordingly, new logistics organisation is appealing. New developments in logistics organisations began to emerge making logistics more open, collaborative, and interoperable. This evolution required the sharing of logistical resources and services between companies and decentralised decision-making based on (real-time) data. This type is called an open, intelligent, decentralised logistics system in this thesis. Two reasons have led to this evolution: recent challenges in logistics, i.e. drivers, and recent theoretical and technological developments, i.e. enablers.

Recent challenges in logistics have been driven by evolutions in manufacturing and retailing. From a manufacturing viewpoint, companies are focusing more and more on product innovation (e.g., developing new products and expanding product range) and customisation (products adapted to specific requirements for every customer), in order to rapidly create new markets or to increase their share in a volatile market. From a logistics standpoint, demands for such new products are very difficult to predict as each request may be different and products may require delayed differentiation. In this context, controlling inventory levels is very important but it is difficult to keep levels to a minimum. However, the time to construct such a supply chain and logistics system is short due to the short time-to-market. Competition is, therefore, fierce in terms of time and cost (Shah, 2009).

From a retail and distribution viewpoint, traditional retailers are striving to develop new channels to distribute products, particularly using internet-based solutions, i.e. e-business or e-commerce. This trend was even more striking with the popularisation of the smartphone in 2010. For example, during the so-called Single’s day in China, on 11th November 2016, approximately RMB 120 billion (approximately USD 17 billion) was spent online in 24 hours. Mobile purchases (using smartphones, tablets, etc.) accounted for 82% of purchases, compared to 72% in 2015 and 43% in 20149. This new business approach, called omni-channel retailing, uses a variety of omni-channels providing the customer with a fully integrated shopping experience (Dholakia et al., 2005; Verhoef et al., 2015), see Figure 7. It has a significant impact on traditional distribution methods for both manufacturers and retailers. Most of them have to shift from B2B (business to business) to B2C (business to consumer) or to B2b2C (Business to small Business to Consumer), for example. From a logistics standpoint, the distribution chain is shorter so manufacturers or retailers are closer to end consumers. The delivery time - from online payment to delivery to the consumer - is crucial to the success of such a business approach. Same-day delivery, or delivery within a few hours, is becoming one of the most valuable services for e-shoppers. However, fast delivery means that e-retailers must accept cost and reliability constraints (the case of Amazon in Figure 4, for example). In addition, omni-channel retailing also tests the flexibility of the logistics system. Firstly, e-retailers have to cope with highly variable demands, especially due to seasonality and (online) promotions (e.g., cyber Monday, single’s day). As current logistics systems often have a fixed warehouse and transport capacity, the variation in demand may

result in unused capacity or unmet demands10. Secondly, e-retailers are expanding delivery channels and methods to satisfy the consumer’s shopping experience and/or to reduce delivery costs. Examples include home delivery, click & collect to stores, pick-up points, collection from depot/warehouse, etc., see Lowe and Rigby (2014). Assuming that purchasers can freely select one of these delivery options, logistics systems have to be more responsive and flexible, i.e. more agile logistics, to cope with unpredictable requests. Therefore, all these new challenges stemming from manufacturing or retailing are compelling logistics to be agile, cost-effective, reliable, and sustainable.

Figure 7. Omni-channel retailing and logistics11

Regarding enablers, recent theoretical and technological developments and innovations have been stimulating advances in logistics more than ever. Enablers can generally be divided into three categories: new theories, new technologies and new techniques. New theories provide new concepts and material to enable innovation in logistics. Examples include the sharing economy (e.g., warehouse sharing such as www.flexe.com), the circular economy (Genovese et al., 2017), crowd shipping (Chen et al., 2017; Mehmann et al., 2015), and horizontal collaboration (e.g., Type 3). Researchers and practitioners rely on these theories to develop new concepts for more efficient logistics systems. New technologies, including 3D printers, Internet of Things (IoT), ICT, cloud computing, robotics, drones, self-driving vehicles, virtual reality (VR), augmented reality (AR), etc., can effectively solve technological issues regarding new logistics organisations such as how to capture real-time information (at object level in particular), how to communicate information for decision-making, and how to rapidly execute decisions. New techniques are for optimisation and decision-making, e.g., big data analytics, artificial intelligence, complex optimisation methods, and real-time decision-making methods. These techniques could provide adequate support for decision-decision-making involving more dynamic, complex, high-volume information in real-time.

As an example, robotisation and automation is a current, rapidly evolving field and it has already been proved that robotisation and automation may help to improve efficiency and effectiveness in logistics. For example, the Roland Berger Study 201512 predicts that, for some positions in logistics, robots will soon be cheaper than human operators, which is also why many companies are now focusing on developing automated sorting systems (e.g., Intralox), or automated shipping devices (e.g., DHL Parcelcopter and SkyPort13_{, self-driving}

vehicles14). Based on robotisation and automation, we can imagine that logistics planning and decision making could also be improved using other technological and technical enablers. For example, IoT and ICT technologies could collect and communicate real-time information, then data analytics and decision-making models would help real-time decision-making and planning, in contrast to traditional methods. It is conventionally known as intelligent logistics (Crainic et al., 2009; McFarlane et al., 2016).

Furthermore, the intervention and importance of 5PL for future evolutions should be specifically mentioned here. As defined in Table 2, the main task of 5PL is to provide technology- or technique-based innovative solutions to improve SC and logistics performance. Examples include data consulting service providers, IoT solutions providers, and organisations for standardisation of logistics data. Currently, 5PLs already play an import role in logistics. For example, IT, such as RFID for freight traceability, IT-based WMS (warehouse management system), and ICT or EDI for collaboration along or across supply chains, is often used to improve logistics efficiency. Considering an open, intelligent, decentralised logistics environment, 5PL is likely to become increasingly important.

From the aforementioned examples, we can see that recent theoretical and technological developments and innovations contribute to the evolution of logistics systems towards open, intelligent, decentralised systems, which can be briefly described as follows:

§ Open means users (LSP, shippers, receivers) can join or leave the logistics network relatively easily. The system hinges on high interoperability between LSPs enabled by standardised materials and procedures thus enabling its capacity to be reduced or extended by simply plugging in or unplugging a service. More importantly, thanks to its openness, on-demand transport or storage services can mitigate problems relating to flexibility and agility in today’s logistics.

§ Intelligent means the logistics system is able to forecast subsequent requests for logistics services, then optimise and plan in advance the short-term logistics (McFarlane et al., 2016). This can be achieved using asset-based intelligence, e.g., intelligent containers (Sallez et al., 2016), or organiser-based intelligence, e.g., business intelligence. We assume that intelligence will probably rely on data-based technologies and techniques such as IoT, big data analytics, or machine learning.

§ Decentralised means decisions in logistics systems are made or updated according to local real-time information instead of centralized advanced planning (Sternberg and Andersson, 2014). It is considered an effective solution to cope with the high dynamic complexity of logistics today. For example, decentralised planning has been investigated in inventory management (Andersson and Marklund, 2000; Shao et al., 2011; Zinn et al., 1989) or in freight transport (Sarraj et al., 2014a; Sternberg and Andersson, 2014). Sometimes,

12 _{https://www.rolandberger.com/publications/publication_pdf/of_robots_and_men___in_logistics.pdf} 13 _{http://www.dpdhl.com/en/media_relations/specials/parcelcopter.html}

decentralisation and intelligence are used together for decision making, namely decentralised intelligence in Sternberg and Andersson (2014).

In accordance with recent evolutions in logistics, in this thesis we will propose and discuss a possible scenario for a future logistics system, namely a self-organised logistics system. It can generally be described as “an open, intelligent, holonic logistics system that aims to harmonise and guide individuals within the system towards a system-wide common goal without significant human intervention from outside” (Pan et al., 2017c). This system will be discussed in more detail in Chapter 4.

1.5 Limitations and Scope of This Habilitation Thesis

According to the objective of this document, as well as to limit the scope of discussion, not all the logistics activities and organisations introduced above are addressed. The choices made and the reasoning are as follows:

§ Logistics activities: this thesis focuses on freight transport. However, some relevant topics such as ICT, business models or new inventory management models (since re-organising freight transport will have an impact on inventory management) will also be discussed. This choice was made for two reasons. Firstly, transport is an important activity in logistics, especially with regard to sustainability, as discussed above. Secondly, my research activities primarily address freight transport.

§ Logistics objectives: as stated in Section 1.2.3, this document mainly discusses

sustainability. It argues that considering sustainability as an objective in addition to

the others would fundamentally test current logistics organisations. Hence, further evolutions are appealing.

§ Logistics organisations: only Type 3 and Type 4 presented in Figure 6 are discussed in the rest of this thesis so as to focus more on the bleeding edge of research in logistics. It is also because of my own research interest, which is heavily focused on Type 3, as well as the desire to move from Type 3 to Type 4 logistics organisations. More specifically, in this thesis, we will firstly review the literature related to Type 3 horizontal collaborative and interconnected logistics, as it can be seen as the most significant evolution in logistics over the past ten years. Academic and practical contributions to this type over the past decade will also be analysed. The comprehensive literature review will help us to identify key research issues and scientific contributions from the international community and thus enable us to position and discuss our research and contributions. Furthermore, the discussion regarding research prospects will focus on type 4 open, intelligent, decentralised logistics, as it can be seen as the future line of research.

After this introductive chapter, the next chapter will focus on Type 3 Horizontal collaborative and interconnected logistics. We have skipped directly to this type (without taking a close look at Types 1 and 2) for two reasons. Firstly, horizontal collaboration has been the most important and noteworthy evolution in logistics over the past ten years. Secondly, my personal contribution to the field of logistics also started with this type. To help the discussion, we start with an in-depth literature review. Through an analytical positioning framework, we will discuss recent international scientific contributions to this type.

C

HAPTER

2

Horizontal Collaborative and Interconnected Logistics:

State of the art

2.1 Chapter Introduction

Being aligned with Figure 6 in Chapter 1, from this chapter we will discuss research works devoted to Type 3 - Horizontal collaborative and interconnected logistics. Since the organisation represents the most important and noteworthy evolution in the last ten-year history of logistics - the collaboration extended from vertical to horizontal. In particular, freight transport, being one of the most significant activity to sustainability in logistics, has attracted increasing attention within this organisation. For those reasons, this chapter will focus on freight transport in the context of horizontal collaborative and interconnected logistics. To simplify, we use in this chapter the term “horizontal collaborative transport” (or HCT) for short. It aims to analyse the state of the art and the international scientific contributions to the research topic. A framework is developed to position the most relevant and important research works. By that, we aim to discover the most important research topics and contributions, as well as gaps in the current state-of-the-art.

This chapter is based on a paper (Pan et al., 2017b) submitted to International Journal of Physical Distribution & Logistics Management. It is organised as following. First, the context and motivation of the research will be presented. Then, Part 3 introduces a two-axis framework of analyse, one axis of solutions and another of scientific issues. Representative and significant scientific publications will be positioned and discussed under the framework. Part 4 will discuss some key findings from the survey; and Part 5 to discuss further research prospects relevant. Finally, Part 6 d will conclude this chapter.

2.2 Horizontal Collaborative Transport and Sustainability

Over the past decades, sustainability in freight transport has become a major preoccupation in the field of logistics (McKinnon et al., 2015; Touboulic and Walker, 2015). On the one hand, freight transport is a lever for economic growth and on the other hand, it contributes significantly to problems such as CO2 emissions, road accidents, and congestion in many

countries (Goldsby et al., 2014; Piecyk and McKinnon, 2010). More effective and efficient organisation of freight transport has become crucial to succeed in logistics, as well as to deal with sustainability challenges.

Among other solutions for sustainable transport such as new fuels or engines, eco-driving, new regulations, electric vehicle, new logistics schemes, as reported in European Commission (1998), collaboration between logistics parties has been recognised as one of the most effective approaches to improve freight transport efficiency and sustainability at logistics level (Goldsby et al., 2014). Although cooperation, coordination, and collaboration should be differentiated (the partnerships vary from operational to strategic level (Spekman et al., 1998)), the term “collaboration” is used here to broadly cover collaborative partnerships in transport and logistics from operational level to strategic level. As discussed in Mason et al. (2007), there are two types of collaboration in logistics: vertical collaboration (VC) and horizontal collaboration (HC). VC focuses on the beneficial vertical relationships between

parties within a supply chain (SC). It has been widely studied in the literature and several surveys are already available, for example Barratt (2004), Power (2005) and Stadtler (2009). HC is more recent and concerns collaboration across SCs. Generally speaking, HC refers to “active cooperation between two or more firms that operate at the same level of the supply chain…”, between shippers, between LSPs, or between receivers for example (Cruijssen, 2006; Mason et al., 2007). Some successful examples of HC can be found in passenger transport with airline alliances such as Skyteam and Star Alliance.

Specifically, in the field of freight transport, the development of HC can be observed from both industrial and academic viewpoints. From an industrial viewpoint, HC is not a new concept, especially in the trucking industry. Well-known examples include the European carrier association ASTRE (ASTRE, 2016) that was created to help independent carriers exchange transport requests. Another example is the well-known collaborative practice conference in liner shipping (Agarwal and Ergun, 2010). Over the past decade, companies have been looking for greater synergy in freight transport to mitigate the higher pressure of logistics costs and demanding services (Cruijssen, 2006). As a result, collaboration across independent SCs to consolidate flows, which is more extensive and efficient than simple lane exchange between carriers, has been considered as an innovative freight transport and logistics solution (Cruijssen, 2006; Mason et al., 2007; Schmoltzi and Wallenburg, 2011). The case of four collaborating manufacturers in France (Mars, UB, Wrigley and Saupiquet) is one of the success stories (CO3, 2014). More industrial case studies can be found in Saenz et al. (2015). From an academic viewpoint, HC is a relatively young but rapidly evolving stream for which new lines of research can be observed. In recent years, a number of relevant concepts, methods, and models have been initiated and studied (see Section 4). These contributions promote HC in freight transport from carrier to SC level, and more recently at supply network level.

Considering the importance and rapid development of HC in freight transport, this chapter focuses exclusively on this topic, which is termed herein “horizontal collaborative transport” (HCT). As there is currently no collective definition of HCT, this document offers a broad, generic one: HCT refers to all types of horizontal collaboration in freight transport between parties operating at the same level of the supply chain (carriers, logistics service providers or shippers), between independent supply chains, and between transport networks, from occasional cooperation to long-lasting collaboration, and from operational level to strategic level. This definition is broad enough to cover the related literature concerning cooperation, coordination, and collaboration in freight transport, and to help provide us with an exhaustive review of the domain. The definition also suggests that HCT can be achieved from various perspectives from transport to supply chain level, and from operational to strategic level. In the survey, we use the term “HCT solutions” to cover all concepts, methods, and models aimed at achieving HCT. Afterwards, the key implementation issues of the solutions will be discussed.

2.3 Analysis of Research Contributions

To provide a structured review of the research on HCT, relevant research works are classified according to two axes: HCT solutions and implementation issues addressed. The classification is proposed to answer to two significant research questions related to the topic. Frist, from industrial perspectives, how should logistics companies (e.g., carriers, LSP, shippers, receivers) adopt effective and efficient HCT solutions, by taking into account their position,

resources, and responsibility in the SC, while being aware of the underlying issues and difficulties when implementing the solutions (called implementation issues in this paper)? Second, from academic perspectives, what are the recent research trends and gaps in HCT, particularly regarding each solution and each issue?

The first axis of classification is called HCT solutions. All HCT concepts, methods, and models reported in the papers reviewed could be classified into six classes of solutions: Single carrier collaboration (S1); Carrier Alliance/Coalition (S2); Transport Marketplace (S3); Shipper or LSP collaboration (S4); Logistics pooling (S5); Physical Internet (S6).

The second axis of classification focuses on the implementation issues involved in HCT solutions. Overall, seven classes could be observed from the literature: Collaborative network design (I1); Transport planning optimisation (I2); Mechanism for exchanging requests (I3); Coalition formation and Gain sharing (I4); Information and communications technology (I5); Organisation (I6); Management and governance (I7); Collaborative and Distributed Inventory Management (I8).

2.3.1 Classification of HCT solutions

This section discusses one by one the definitions and collaboration schemes of the six classes of HCT solutions introduced previously.

2.3.1.1 Single carrier collaboration (S1)

Single carrier collaboration is a term used in Hernández et al. (2011) to describe an HCT solution for an autonomous and independent carrier who collaborates with one or more other carriers. It is a bilateral collaboration between carriers motivated by at least three goals, i.e., reduce transport costs (Hernández et al., 2011), acquire external capacity to serve excess requests (Hernández and Peeta, 2014), or improve services for the (same) client (Puettmann and Stadtler, 2010). Practical examples include collaboration between express carriers, for example, Fedex (in the US) and Chronopost (in France) collaborating to improve local delivery services and efficiency in both countries.

Figure 8. Single carrier collaboration scheme

Figure 8 illustrates the single carrier collaboration scheme. For each SC, shippers will procure transport services from a carrier to ship freight to a receiver. The two carriers serving different SC may exchange on hand requests in order to improve transport efficiency and thus profitability. Moreover, a carrier takes a request from another with the transport constraints given by the shipper and/or the receiver. The constraints can be lane, volume, lead time, delivery time windows, etc. The service undertaken must be maintained in the two SC after the exchange.

Exchanges

Shipper 1 Carrier 1 Receiver 1